Archaeometric evidence for the earliest exploitation of lignite from the bronze age Eastern Mediterranean Stephen Buckley1,2, Robert C. Power3,4, Maria Andreadaki-Vlazaki5, Murat Akar6, Julia Becher1,2, Matthias Belser1, Sara Cafisso1, Stefanie Eisenmann4, Joann Fletcher7, Michael Francken8, Birgitta Hallager9, Katerina Harvati8, Tara Ingman10, Efthymia Kataki11, Joseph Maran12, Mario A. S. Martin13,14, Photini J. P. McGeorge15, Ianir Milevski16, Alkestis Papadimitriou17, Eftychia Protopapadaki11, Domingo C. Salazar-García18,19, Tyede Schmidt-Schultz20, Verena J. Schuenemann8,21, Rula Shafiq22, Ingelise Stuijts23, Dmitry Yegorov16, K. Aslιhan Yener24, Michael Schultz20,25, Cynthianne Spiteri1 & Philipp W. Stockhammer3,4 Scientific Reports 11, Article number: 24185 (2021) Cite this article 745 Accesses 76 Altmetric Metrics BiogeochemistryMass spectrometry This paper presents the earliest evidence for the exploitation of lignite (brown coal) in Europe and sheds new light on the use of combustion fuel sources in the 2nd millennium BCE Eastern Mediterranean. We applied Thermal Desorption/Pyrolysis–Gas Chromatography-Mass Spectrometry and Polarizing Microscopy to the dental calculus of 67 individuals and we identified clear evidence for combustion markers embedded within this calculus. In contrast to the scant evidence for combustion markers within the calculus samples from Egypt, all other individuals show the inhalation of smoke from fires burning wood identified as Pinaceae, in addition to hardwood, such as oak and olive, and/or dung. Importantly, individuals from the Palatial Period at the Mycenaean citadel of Tiryns and the Cretan harbour site of Chania also show the inhalation of fire-smoke from lignite, consistent with the chemical signature of sources in the northwestern Peloponnese and Western Crete respectively. This first evidence for lignite exploitation was likely connected to and at the same time enabled Late Bronze Age Aegean metal and pottery production, significantly by both male and female individuals. Making fire was a crucial stage in the development of humankind1,2 and inhaling its smoke was an inevitable result of this process. The increasing ability to manage fires and their temperatures also allowed for more sophisticated methods of cooking, and enhanced the development of craft technologies such as pottery production and the melting and casting of metals. To date, ancient fire-making has mostly been studied through the residual ash and other remains taken from fireplaces, and from the thermal alteration and soot marks visible on artifacts, in addition to the information provided by experimental archaeology and ethnoarchaeology (e.g.3,4,5). Only recently has it become possible to understand the inhalation of smoke by past individuals by studying its chemical and microscopic traces in human dental calculus6,7,8,9. Our study of combustion markers in human dental calculus is part of a larger project which aims to understand culinary practices of the 2nd millennium BCE Eastern Mediterranean, i.e. the Middle Bronze Age (ca. 2000–1600 BCE), the Late Bronze Age (ca. 1600–1200/1050 BCE) and the Early Iron Age (after 1200/1050 BCE) (Fig. 1; Table S1; SM Text 2). During this time, the Eastern Mediterranean transformed into an early globalized region, characterized by complex stratified societies employing writing systems and sophisticated craftsmanship as well as large-scale production centres producing goods often aimed at trans-regional exchange. The written and archaeological evidence also reveals a high degree of individual mobility between the Mycenaean Aegean, Hittite Anatolia, Cyprus, the Levant’s trade centres and city kingdoms, Mesopotamia and Middle and New Kingdom Egypt10,11,12. Therefore, we have selected human dental calculus from key sites of the 2nd millennium in the Aegean (Tiryns, Chania), the Levant from the north (Alalakh) via present-day Lebanon (Kamid el-Loz) to the south (Megiddo, Tel Erani) and Egypt (Abusir el-Meleq, Thebes). Tiryns and Chania were major harbour and palatial sites in the Aegean. Tiryns was a focus of large-scale craft production of the Northeastern Peloponnese and closely linked to Mycenae, the nearby major Late Bronze Age political centre in that region of Southern Greece13. As part of the increasing control of Crete by the Mainland palaces, Chania became one of the key centres for the Mainland’s exploitation of the island14. The Levantine cities (Alalakh, Kamid el-Loz, Megiddo, Tel Erani) were situated in between and usually influenced by or under the control of the great empires of Anatolia, Syro-Mesopotamia and Egypt, the latter represented by samples in our study. The 2nd millennium was also marked by the large-scale production of goods and their subsequent trade, especially pottery and metal objects, throughout the Eastern Mediterranean10. Most noteworthy is the mass production of pottery in the Late Bronze Aegean, a large part of which was exclusively produced for export to Cyprus and the Levant. So far, there has been little discussion of the resources that were used to fuel the kilns and ovens for this proto-industrial production in densely settled and probably largely de-forested areas, whereas the fuels used for cooking have been increasingly studied in the Eastern Mediterranean Bronze Age4,15,16,17. Our study of chemical combustion markers aims at a better understanding how this unprecedented level of interconnection transformed both local cooking practices and the procurement of fuel.
Figure 1Map of the Eastern Mediterranean featuring the sites included in this study and the currently-known key lignite sources in Greece (created by R.C.P. with QGIS3, using Natural Earth raster data: QGIS.org, 2021. QGIS Geographic Information System. QGIS Association. http://www.qgis.org).Thermal Desorption-Gas Chromatography-Mass Spectrometry (TD-GC–MS) and Pyrolysis–Gas Chromatography-Mass Spectrometry (Py-GC–MS) of the human dental calculus revealed a significant abundance of combustion markers in 74 of the 77 samples retrieved from the 67 individuals studied (SM Appendix, Text 1.2; Tables S1, S2a and S2b).Wood and dung combustionThe dominant biomarkers identified were polynuclear aromatic hydrocarbons (PAHs), i.e. organic compounds with two to six membered aromatic rings typical of chars or soot associated with smoke resulting from the long-term and/or repeated exposure to fires in more or less close vicinity. The ratios of these PAHs are typical of combustion, rather than petrogenic sources deriving from modern environmental contamination18,19,20, as would be expected for ancient exposure to smoke/fires. All samples with significant abundances of PAHs also revealed biomarkers indicative of conifer wood combustion. These included the diterpenoid acid esters methyl dehydroabietate and methyl abietate, their free acids being significant components found in Pinaceae wood (which includes genera such as Pinus, Picea, Larix, Abies and Cedrus) and resin. These thermally-derived methyl esters were accompanied by defunctionalized diterpenoids which included 19-norabieta-4,8,11,13-tetraene, 19-norabieta-4(18),8,11,13-tetraene, 19-norabieta-3,8,11,13-tetraene, tetrahydroretene, dihydroretene, retene and dehydroretene. These methyl esters and defunctionalized diterpenoids reflect the strong heating process which the conifer wood would have undergone during its burning, and its production of the subsequently inhaled smoke. Additional chemical evidence for conifer wood combustion is revealed by the relative abundance of the dimethylphenanthrenes (DMPs). These have been shown to be diagnostic, with 1,7-dimethylphenanthrene, deriving chemically from the diterpenoids characteristic of conifers, dominating the m/z 206 profile and producing distinctively high 1,7/1,7 + 2,6-DMP ratios of ~ 0.921,22. The presence of these biomarkers and the characteristic ratios confirms the inhalation of smoke from the burning of conifer (e.g. Pinaceae) woods23,24,25. More generally, the ratios of the PAHs fluoranthene to pyrene have allowed us to further identify and discriminate between dung, wood, and fossil fuel sources, with ratios of ~ 0.8 for dung, ~ 1 for wood and 1.40 for lignite/coal26,27. A 1,7/1,7 + 2,6-DMP ratio of ~ 0.5 is typical of animal dung burning, whereas values for oakwood (hardwood) and pinewood (softwood) are ~ 0.7 and ~ 0.9 respectively18. The use of dung as fuel was attested in 29 of the individuals studied here, based on the combined 1,7/1,7 + 2,6-DMP ratio consistent with the burning of dung (see SI 1), and a fluoranthene/pyrene ratio of ~ 0.8 indicative of the combustion of dung26,27 (Table S1). Hardwoods and softwoods were also used as fuels, the chemical data identifying their use in 58 and 64 individuals respectively (Table S1), and indicating exposure to smoke from multiple fuel sources in the case of the majority of individuals, presumably reflecting local availability of potential fuel resources at these sites.Lignite (brown coal) combustionAn unusual biomarker distribution (Fig. 2) was observed in six Late Bronze Age and two Early Iron Age individuals from Tiryns and three Late Bronze Age individuals from Chania (Table S1). The most diagnostic components included succinimide (m/z 56, 99), benzoic acid (m/z 77, 105, 122), benzamide (m/z 77, 105, 121), phthalic anhydride (m/z 76, 104, 148) and phthalimide (m/z 76, 104, 147) (Fig. 2 and Table S2b). The finding of the succinimide and aromatic biomarkers in samples from Tiryns and Chania and, thus, only in samples from the Aegean, is significant. These specific biomarkers are usually observed together in self-ignited brown and black coal heaps28,29 where the potential for self-combustion is high. The absence of these characteristic aromatic biomarkers from the vast majority of the calculus samples, and their co-occurrence specifically in the PAH-rich calculus samples, excludes exogenous, pre- or post-excavation, contamination as a possible source. The presence of these biomarkers is particularly significant since Greece has important lignite deposits, all of which are surface deposits mined as opencast resources30,31,32,33,34, which would also have made them visible to ancient populations. The main lignite deposits were formed in intermountain basins such as Ptolemais in Macedonia and Megalopolis in the central Peloponnese, while smaller lignite deposits were created in the western Peloponnese at sites such as Olympia and Pyrgos30,31 as well as in Kandanos and Vrysses Apokoronou in western Crete33,34. Lignite chemistry reflects the plant input of their formation, which can thereby provide clues to potential sources. The chemical profiles in the calculus samples containing the lignite biomarkers from both Tiryns and Chania do not reveal chemical evidence for sulfur or its derivatives, suggesting the lignite had a low sulfur source. Megalopolis is the closest lignite deposit to Tiryns, although lignite from this site is known to be relatively high in sulfur31,32 which is inconsistent with the evidence from Tiryns. Further investigation of the full-scan TD–GC–MS data from the Tiryns calculus samples containing the lignite-related biomarkers also revealed an unresolved complex mixture (UCM), observed as a notable hump in the chromatograms and typical of lignites35. Furthermore the biomarkers present are characteristic of higher plants from both gymnosperms and angiosperms consistent with the palaeoecology of southeastern Europe36. The gymnosperm-derived diterpenoid compounds dominate the higher plant input, and include not only the fossil biomarkers dehydroabietane, simonellite and 2-methylretene, which can derive from a wide range of conifers, and the more specific 6-dehydroferruginol and diaromatic totarane indicative of a Cupressaceae (e.g. Taxodium sp., subfamily Taxodioideae)37,38,39,40. Importantly, there were no pimarane- or isopimarane-type diterpenoids, nor abietane-type diterpenoid acids detected, which suggests that a significant pine (Pinus spp.) input for the lignite can be excluded37,40,41. Notably, the diterpenoids identified are also accompanied by a number of monoterpenoids: p-cymene, thujone and carvenone, and the sesquiterpenoids: calamenene, α-calacorene, calamene, cadalene and isocadalene. These terpenoid counterparts to the particular diterpenoids identified have been observed in several Cupressaceae, in particular Taxodium spp.39,42,43. Lignite from Megalopolis has been noted for its lack of gymnosperms32 and an angiosperm predominance32,36, which again argues against this as the likely lignite source. The relative abundance of these mono-, sesqui- and diterpenoids is interesting, given the prevalence of conifers growing in a swamp environment, typical of Taxodium species and other closely related Cupressaceae (and subfamily Taxodioideae) known to have grown in the western Peloponnese during the late Miocene Epoch36. Present-day Taxodium and other Taxodioideae are not found in Europe and only in North America. The biomarker evidence for angiosperms includes the presence of 18α(H)-oleanane and it is notable that sample TIR002B shows an n-alkane Cmax at C29 for n-nonacosane, which is consistent with and indicative of Quercus (oak) species, C29 known to be the dominant n-alkane in Quercus38. Additionally, the more diagnostic angiosperm biomarkers3,3,7,12a-tetramethyl-1,2,3,4,4a,11,12,12a-octahydrochrysene and 2,2,4ab,9-tetramethyl-1,2,3,4,4a,5,6,14b-octahydropicene44, were identified and have been found in angiosperm-containing ligni
https://www.nature.com/articles/s41598-021-03544-w
